Conventionally, in order to improve the level of freedom of cell architecture in cellular systems, configurations in which the functions of a base station apparatus are divided between a signal processing unit (hereinafter referred to as “BBU” (BaseBand Unit)) and an RF unit (hereinafter referred to as “RRH” (Remote Radio Head)), and the BBU and the RRH are physically separated, have been considered. In such a configuration, the wireless signals transmitted between the BBU and the RRH are transmitted by means of RoF technology. RoF technologies can be largely divided between analog RoF technologies and digital RoF technologies, depending on the optical transmission method. In recent years, the study of digital RoF technologies having superior transmission quality has flourished, and standardization organizations such as the CPRI (Common Public Radio Interface) and the like are working towards the establishment of specifications (see, e.g., Non-patent Document 1). Additionally, while coaxial cable, optical fiber, and the like can be used as the connecting media between BBUs and RRHs, the transmission distance can be extended, in particular, by using optical fiber to connect BBUs and RRHs.
Herebelow, digital RoF transmission will be explained.
In discussing digital RoF transmission, the following terminology will be defined.
A downlink refers to the communication path of radio waves transmitted from a BBU, via an RRH, to a wireless terminal connected to the RRH.
An uplink refers to the communication path of radio waves transmitted from a wireless terminal connected to an RRH, via the RRH, to a BBU.
On a digital RoF transmission downlink, the following processes are performed. A BBU prepares a digital signal (hereinafter referred to as “IQ data”) separately for the I-axis and Q-axis components of a wireless signal, converts the prepared IQ data into an optical signal, and transmits the converted optical signal to an RRH via an optical fiber. The RRH converts the received optical signal into a wireless signal, and transmits the converted wireless signal to a wireless terminal.
Additionally, on a digital RoF transmission uplink, the following processes are performed. An RRH receives a wireless signal transmitted from a wireless terminal, converts the received wireless signal into an optical signal, and transmits the converted optical signal to a BBU via an optical fiber. The BBU converts the received optical signal into IQ data and demodulates the signal.
FIG. 17 is a schematic block diagram illustrating the functional structure of an RRH 500 during digital RoF transmission.
The RRH 500 includes an antenna 501, a transmission/reception switching unit 502, an amplifier 503, a down-conversion unit 504, an A/D (Analog/Digital) conversion unit 505, a baseband filter unit 506, a framing unit 507, an E/O (Electric/Optic) conversion unit 508, an O/E (Optic/Electric) conversion unit 509, a deframing unit 510, a baseband filter unit 511, a D/A (Digital/Analog) conversion unit 512, an up-conversion unit 513, and an amplifier 514.
The antenna 501 transmits and receives wireless signals. The transmission/reception switching unit 502 switches the antenna 501 between transmission and reception. The amplifier 503 amplifies the signal power of a received wireless signal to a level that allows for signal processing. The down-conversion unit 504 down-converts the amplified wireless signal into the baseband. The A/D conversion unit 505 converts the down-converted wireless signal (analog signal) into IQ data, which is a digital signal. The baseband filter unit 506 performs a filtering process on the IQ data. The framing unit 507 performs framing by multiplexing the filtered IQ data with a control signal. The E/O conversion unit 508 converts the framed signal (hereinafter referred to as the “frame signal”) (electrical signal) into an optical signal, and transmits the converted optical signal to the BBU via an optical fiber 550.
The O/E conversion unit 509 converts an optical signal received via the optical fiber 550 into a frame signal (electrical signal). The deframing unit 510 extracts a control signal and IQ data from the frame signal. The baseband filter unit 511 performs a filtering process on the IQ data. The D/A conversion unit 512 converts the filtered IQ data into an analog signal. The up-conversion unit 513 up-converts the analog signal. The amplifier 514 amplifies the power of the analog signal to a predetermined transmission power.
FIG. 18 is a schematic block diagram illustrating the functional structure of a BBU 600 during digital RoF transmission.
The BBU 600 includes an O/E conversion unit 601, a deframing unit 602, a modulation/demodulation unit 603, a framing unit 604, and an E/O conversion unit 605.
The O/E conversion unit 601 converts an optical signal received via an optical fiber 650 into a frame signal (electrical signal). The deframing unit 602 extracts a control signal and IQ data from the frame signal. The modulation/demodulation unit 603 restores a wireless signal by demodulating the IQ data. Additionally, the modulation/demodulation unit 603 outputs the IQ data of the wireless signal to the framing unit 604. The framing unit 604 performs framing by multiplexing the IQ data with a control signal. The E/O conversion unit 605 converts the frame signal (electrical signal) into an optical signal and transmits the converted optical signal to the RRH 500 via the optical fiber 650.
Digital RoF transmission requires an extremely broad band in the optical fiber section. For example, in an LTE (Long Term Evolution) system, the wireless signals in a 2×2 MIMO (Multiple-Input and Multiple-Output) with a system bandwidth of 20 MHz have a maximum data rate of 150 Mbps in the wireless section. However, in order to transmit these wireless signals at a 15-bit quantization bit rate, a CPRI link of option 3 (2.4576 Gbps) or greater is needed. Therefore, the application of compression technologies to digital RoF transmission is being studied in order to make effective use of the optical band. Compression techniques can be largely divided between lossy compression and lossless compression. Lossy compression includes reduction of the sampling frequency, reduction of the quantization bit rate, or the like. Lossless compression includes a combination of linear predictive coding and entropy coding or the like. For example, when raising the transmission rate in the wireless section, the required transmission band in the optical section will also increase, but the increased speed in the wireless section can be handled without changing the optical transceiver if the required transmission band in the optical section is reduced by compression technology. For example, Non-patent Document 2 discusses MPEG-4 ALS (Moving Picture Experts Group-4 Audio Lossless Coding), which is a lossless compression technique.
FIG. 19 is a schematic block diagram illustrating the functional structure of an RRH 500a when incorporating compression technology during multiplexed transmission.
The RRH 500a includes an antenna 501, a transmission/reception switching unit 502, an amplifier 503, a down-conversion unit 504, an A/D conversion unit 505, a baseband filter unit 506, a compression unit 701, a framing unit 507a, an E/O conversion unit 508, an O/E conversion unit 509, a deframing unit 510, an expansion unit 702, a baseband filter unit 511a, a D/A conversion unit 512, an up-conversion unit 513, and an amplifier 514.
The compression unit 701 compresses filtered IQ data. The framing unit 507a performs framing by multiplexing the compressed IQ data with a control signal. The expansion unit 702 restores the IQ data by decompressing the compressed IQ data. The baseband filter unit 511a performs a filtering process on the restored IQ data. FIG. 20 is a schematic block diagram illustrating the functional structure of a BBU 600a when incorporating compression technology during multiplexed transmission.
The BBU 600a includes an O/E conversion unit 601, a deframing unit 602, an expansion unit 801, a modulation/demodulation unit 603a, a compression unit 802, a framing unit 604a, and an E/O conversion unit 605.
The expansion unit 801 restores IQ data by decompressing compressed IQ data. The modulation/demodulation unit 603a restores a wireless signal by demodulating the restored IQ data. Additionally, the modulation/demodulation unit 603a outputs the IQ data of the wireless signal to the compression unit 802. The compression unit 802 compresses the IQ data. The framing unit 604a performs framing by multiplexing the compressed IQ data with a control signal.
Among compression technologies, there are those in which a compression process and an expansion process are performed for every predetermined number of samples. In the following explanation, the units for performing the compression process will be referred to as frames, and the predetermined number of samples will be referred to as the frame size. For example, in compression technologies using linear predictive coding, a predicted value is obtained by multiplying coefficients by a number of sample points that are older than a given sample point and adding the multiplication results, and the error between the predicted value and the given sample point is outputted. If the prediction accuracy is high, then the amplitude value of the error signal will be close to zero. For this reason, the required band in the optical section can be reduced by entropy coding for transmitting data at a lower bit rate for amplitude values near zero, which have a higher probability of occurrence. It is to be noted that the coefficients are determined separately for each frame, and calculated so that the prediction error will be small for the IQ data in each frame.
Next, LTE wireless signals will be explained.
In LTE, OFDM (Orthogonal Frequency Division Multiplexing) is used in the downlink. As the time waveform, a signal having a cyclic prefix appended to a signal of a predetermined size that has been subjected to an IFFT (Inverse Fast Fourier Transform) is periodically outputted. On the other hand, in LTE, DFT-S-OFDM (Discrete Fourier Transform-Spread-OFDM) is used in the uplink. In this case also, as with OFDM, a signal having a cyclic prefix appended to a signal of a predetermined size that has been subjected to an IFFT is periodically outputted as the time waveform. In the following explanation, a signal having a cyclic prefix appended to a signal that has been subjected to an IFFT will be referred to as an OFDM symbol, without making a distinction between the downlink and the uplink.
In LTE, a normal cyclic prefix and an extended cyclic prefix are defined. A normal cyclic prefix is shorter than an extended cyclic prefix, and has higher frequency utilization efficiency. For this reason, normal cyclic prefixes are normally used, and in the following description, normal cyclic prefixes will be explained as an example. FIG. 21 illustrates the structure of time slots in LTE. In the example shown in FIG. 21, seven OFDM symbols are arranged in a 0.5 ms interval. If the system bandwidth is 20 MHz, then the IFFT size is 2048, the size of the cyclic prefix (CP1) of the first OFDM symbol is 160 points and the size of the cyclic prefix (CP2) of the second to seventh OFDM symbols is 144 points. Therefore, the OFDM symbol length is 2208 points for the first OFDM symbol and 2192 points for the second to seventh OFDM symbols. Thus, the OFDM symbol lengths are not all the same. Non-patent Document 3 describes the structure of LTE frames.
FIG. 22 is a diagram illustrating the compression rate for each frame when applying MPEG4-ALS to the data in the I component of a wireless signal.
In FIG. 22, the frame number represents the order in which the frames were compressed. The compression rate is the ratio of the data amount after compression to the original data amount. The wireless signal was OFDM-modulated using 1200 subcarriers with a subcarrier spacing of 15 kHz, modulated by 256 QAM (Quadrature Amplitude Modulation), with cyclic prefixes that were 160 samples (first OFDM symbol) or 144 samples (second to seventh OFDM symbols) long. In other words, it was assumed that the entire wireless band was used for data transmission in an LTE downlink system with a system bandwidth of 20 MHz. The frame size was 548.
In FIG. 22, (a) indicates the compression rate when only the first OFDM symbol is contained in a frame. (b) indicates the compression rate when the first OFDM symbol and the second OFDM symbol are contained in a frame. (c) indicates the compression rate when only the second OFDM symbol is contained in a frame. (d) indicates the compression rate when the second OFDM symbol and the third OFDM symbol are contained in a frame. (e) indicates the compression rate when only the third OFDM symbol is contained in a frame. (f) indicates the compression rate when the third OFDM symbol and the fourth OFDM symbol are contained in a frame. (g) indicates the compression rate when only the fourth OFDM symbol is contained in a frame. (h) indicates the compression rate when the fourth OFDM symbol and the fifth OFDM symbol are contained in a frame. (i) indicates the compression rate when only the fifth OFDM symbol is contained in a frame. (j) indicates the compression rate when the fifth OFDM symbol and the sixth OFDM symbol are contained in a frame.